* * *
Put together, in scrutinizing these standard claims and making these critical arguments, this book seeks to impress upon the reader one main point about military alliances: that these organizations defy easy explanations and are often so puzzling that it should be small wonder that US military alliances – or any alliance for that matter – can at times seem very dysfunctional. States write down their political and military commitments in treaties so as both to clarify their intentions and to create ambiguity over the circumstances in which they would act. Leaders fear that their country will be dragged into disputes they do not wish to have due to the reckless behavior of allied states, but precisely because they have such apprehensions, those fears rarely – if ever – become reality. Worrying about abandonment by an ally is natural and rational, but seldom do these concerns intensify to a level that dramatically reshapes a state’s foreign and defense policy. This may just as well be due to skillful alliance management. Though burden-sharing controversies have dominated many intra-alliance debates since the beginning of the Cold War, they are partly the result of those alliances lasting much longer than ever before, thanks, arguably, to nuclear deterrence, which can create disincentives for states to spend on their militaries. Of course, states do ultimately, though unevenly, build up their military capabilities in order to deter in peacetime and to prevail in wartime, but many, if not most, multinational military campaigns do not involve the full membership of an alliance. They oftentimes include non-allies, which may lead some to ask why it is worth even bothering to have a formal alliance at all. The factor that illuminates why states agree to form a military alliance in the first place often sheds little, if any, light on why that alliance comes to an end. Contradictions are pervasive.
Defining Military Alliances
Before fleshing out these arguments in greater detail, a crucial question remains: what exactly is a military alliance? This question is deceptively simple, not least because news media often invoke the term to describe a wide variety of security arrangements like NATO, bilateral partnerships that involve the United States and countries such as Israel, Saudi Arabia, and Pakistan, as well as the burgeoning cooperation between China and Russia. And indeed, some scholars would agree that such relationships constitute alliances. Michael Barnett and Jack Levy (1991: 370) define an alliance “in its broadest sense to refer to a formal or informal relationship of security cooperation between two or more states and involving mutual expectations of some degree of policy coordination on security issues under certain conditions in the future.” In his seminal study on alliances, Stephen Walt (1987: 12) similarly defines them as “a formal or informal arrangement for security cooperation between two or more sovereign states.” As such, scholars have counted alliances in all sorts of ways, creating confusion as to the true count, and running the risk of comparing apples to oranges. Mira Rapp-Hooper (2020: 17) writes, for example, that the United States had thirty-seven allies as of 2020, but she includes in her count Israel and Pakistan (which do not have a formal defense agreement with the United States) and omits countries that make up the Inter-American Treaty of Reciprocal Assistance, popularly known as the Rio Pact (which notionally does contain one).
For its purposes, this book adopts a more restrictive definition. Specifically, I define military alliances as arrangements made between two or more sovereign states on the basis of a written treaty that serves to coordinate military policy toward at least one common goal. This definition has several components that need unpacking.
First, only sovereign states have military alliances. Sovereignty is, of course, a fraught concept. Few states, if any, are fully sovereign in terms of their own domestic and foreign policies, whether because they delegate some juridical authority to international organizations, subordinate some decision-making to a stronger state, or both (Krasner 1999). In the modern context, though, a country is at least nominally sovereign if it has its own representation at international bodies and, most importantly, is recognized as a state by other states. Of course, violent nonstate actors can align themselves with states or even other nonstate actors so as to have an alliance in the common sense of the term (Tamm 2016; see also Grynaviski 2015). The most famous formal example is the 1778 Treaty of Alliance signed by the Kingdom of France and the Thirteen United States of America during the American Revolutionary War. This example notwithstanding, such relationships are often ad hoc and, as we will see, lack the other key ingredients that make up a military alliance. That said, sovereign states can forge, and have forged, military alliances in order to counter threats emanating from transnational movements.
Second, military alliances are based on some written treaty. This feature is crucial to the definition I use in this book. Treaties outline, with varying degrees of clarity (or, one might say, ambiguity), the rules of the game that constrain their relationship: they spell out the promises and obligations, sometimes even listing the conditions under which their provisions would or would not be activated. Not all written commitments are alike. A mutual defense treaty with reciprocal obligations could constitute the alliance, as in the case of NATO, but a security treaty could simply entail one-sided obligations, as in the case of the US–Japan alliance. The United States is an ally of Thailand even though the founding treaty – the Southeast Asia Collective Defense Treaty (popularly known as the Manila Pact) – produced a multilateral organization that was eventually dissolved in 1977. In addition to the Manila Pact, the 1962 Thanat-Rusk communiqué and the 2012 Joint Vision Statement for the Thai–US Defense Alliance form the basis of their ongoing security partnership.
By stressing the importance of having a written treaty, I exclude informal partnerships or alignments that other scholars like Barnett, Levy, and Walt include in their studies. I also do not cover so-called “alliances of convenience” – that is, instances where adversaries cooperate informally to tackle urgent security challenges (Resnick 2010/11). I believe excluding them is defensible on several grounds. Incorporating both informal (i.e., nontreaty) and formal (i.e., treaty) arrangements in a definition of alliance blurs the distinction between friendly diplomacy and active military partnership. We distinguish between dating and marriage when we discuss romantic relations because they entail different expectations and obligations. We should make similar distinctions with respect to international security cooperation. Treating alignment and alliance as interchangeable introduces unnecessary difficulties in measuring the concept: in the absence of a treaty, how much alignment must we see for it to qualify as a military alliance? The answer to this question is not self-evident. In emphasizing the formal signing of military alliances, I avoid these difficulties. Nevertheless, I am aware that other problems could arise with my decision. The United States is a treaty ally of North Macedonia via NATO, but not of Israel or Saudi Arabia. China and Russia arguably engage in more military cooperation now than some treaty allies have had historically. The reinvigorated Quadrilateral Security Dialogue (the Quad) – comprising Australia, India, Japan, and the United States – does not qualify as an alliance because it is not based on documented security guarantees. Why, then, leave out these more strategically significant relationships from the analysis? Suffice it to say for now, not extending a treaty to such states, however closely aligned they appear to be, is a deliberate choice to avoid the very problems that can come with alliances.
Third, military alliances are focused primarily on coordinating military policy. This point seems redundant, but it is worth highlighting because states cooperate with one another on a wide range of issues. In the economic sphere, they can form free or preferential trade areas, common markets, as well as customs, monetary, or other types of economic unions. They can also participate in other international organizations that center on human rights promotion, environmental protection, or social development. Yet none of these international organizations would qualify as a military alliance because they are not geared toward coordinating military policy – specifically, plans and decisions relating to the kinetic and lethal use of force against third parties. Hence, the United States is not, formally speaking, a military ally of the EU, even though many EU members are also part of NATO. Of course, military alliances do sometimes feature nonmilitary provisions in their charters. Article II of NATO’s founding document – the Washington Treaty – provides that its members “will contribute toward the further development of peaceful and friendly international relations by strengthening